基于条件生成对抗网络的台风极值风速分析

洪旭; 侯子洋; 孔凡; 胡亮

doi:10.7638/kqdlxxb-2024.0213

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于条件生成对抗网络的台风极值风速分析

研究专栏——大型结构风效应流固耦合与人工智能 | 更新时间：2026-05-28

- 基于条件生成对抗网络的台风极值风速分析
- Extreme wind speed analysis of typhoons based on conditional generative adversarial networks
- 空气动力学学报 2025年43卷第5期页码：101-111
- 作者机构：
  
  1. 合肥工业大学土木与水利工程学院,合肥,230009
  2. 圣母大学 Nathaz实验室，南本德,美国,46556
- 作者简介：
  
  洪旭*（1993—），男，安徽黄山人，讲师，研究方向：结构风工程与随机动力学. E-mail：xhong@hfut.edu.cn
- 基金信息：
  
  国家自然科学基金（52408522）;安徽省自然科学基金（2408085QE149）;中央高校基本科研业务费资助项目（JZ2023HGTA0194）
- DOI：10.7638/kqdlxxb-2024.0213
  中图分类号： P444;TU312.1;TP183
- 纸质出版：2025
- 稿件说明：
移动端阅览
洪旭, 侯子洋, 孔凡, 等. 基于条件生成对抗网络的台风极值风速分析[J]. 空气动力学学报, 2025, 43(5): 101−111.

HONG Xu, HOU Ziyang, KONG Fan, et al. Extreme wind speed analysis of typhoons based on conditional generative adversarial networks[J]. 2025, 43(5): 101-111.
洪旭, 侯子洋, 孔凡, 等. 基于条件生成对抗网络的台风极值风速分析[J]. 空气动力学学报, 2025, 43(5): 101−111. DOI： 10.7638/kqdlxxb-2024.0213

HONG Xu, HOU Ziyang, KONG Fan, et al. Extreme wind speed analysis of typhoons based on conditional generative adversarial networks[J]. 2025, 43(5): 101-111. DOI： 10.7638/kqdlxxb-2024.0213.

摘要

为准确评估台风极值风速，保障台风影响地区的重要工程结构抗风安全，建立了基于条件生成对抗网络的台风路径与强度模型，以实现台风极值风速的精细化分析。首先，将台风运动速度、台风强度的6小时变化量处理为一定状态变量（台风中心经纬度等）和环境变量（洋面温度等）的条件随机变量。然后，采用条件生成对抗网络建立台风运动速度、台风强度6小时变化量的生成器和判别器全连接神经网络，利用中国气象局台风最佳路径数据、20th Century Reanalysis（20 CR）再分析数据集等气象资料训练上述神经网络。通过与历史记录的对比，验证了台风路径模型与强度模型能够合理地反映历史台风运动轨迹特征和强度演化规律，并能准确再现台风局部路径中关键参数的均值、标准差乃至概率分布函数等重要统计特征。最后，利用本文台风模型分析了我国东南沿海地区台风极值风速，并与其他研究及规范推荐结果进行对比分析，全面评估了该模型在预测与台风相关的极端风事件方面的准确性和相关性。该模型为了解台风动态提供了更精确的工具，有望帮助改善沿海地区的早期预警系统和风险评估。

Abstract

Accurately assessing the extreme wind speeds of typhoons is of great significance for ensurning the wind-resistant safety of important engineering structures in typhoon-affected areas. A typhoon track and intensity model based on the conditional generative adversarial networks (GAN) to facilitate refined analysis of extreme typhoon wind speeds was developed in the present study. The model began by treating the 6-hour changes in typhoon translational speed and intensity as conditional random variables

which were determined by certain state variables

(e.g.

as the latitude and longitude of the typhoon center) and environmental variables (e.g.

as sea surface temperature). Then

the conditional GAN was applied

where fully connected neural networks were used for both the generator and discriminator to model the 6-hour changes in translational speed and intensity. The neural networks were trained using meteorological data

including the best track data from the China Meteorological Administration and the 20th Century Reanalysis (CR) dataset. The performance of the typhoon track and intensity model was validated through comparisons with historical records

confirming that the models could effectively capture key characteristics of historical typhoon trajectories as well as the evolution of their intensity. The models also demonstrated auurate reproduction of key statistical characteristics in typhoon track patterns

such as mean values

standard deviations

and probability distribution functions for key parameters in localized typhoon paths. Finally

the developed typhoon model was applied to analyze extreme wind speeds along the southeastern coastal regions of China

and the results were compared with other research findings and guideline recommendations

providing a comprehensive evaluation of the model's accuracy and relevance in forecasting extreme wind events associated with typhoons. This model offers a more precise tool for understanding typhoon dynamics

potentially improving early warning systems and risk assessment for coastal regions.

关键词

Keywords

references

GU M, ZHAO M W, QUAN Y. Typhoon risk assessment of structures[J]. Journal of Tongji University (Natural Science), 2009, 37(5): 569−574(in Chinese).

OU J P, DUAN Z D, CHANG L. Typhoon risk analysis for key coastal cities in southeast China[J]. Journal of Natural Disasters, 2002, 11(4): 9−17(in Chinese).

LIN P Y, LIN P H, WANG J, et al. Typhoon disaster risk assessment and dynamic risk forecasts in Zhejiang Province based on machine learning methods[J]. Journal of Natural Disasters, 2023, 32(4): 13−24(in Chinese). doi: 10.13577//j.jnd.2023.0402

VICKERY P J, SKERLJ P F, TWISDALE L A. Simulation of hurricane risk in the U. S. using empirical track model[J]. Journal of Structural Engineering, 2000, 126(10): 1222−1237. doi: 10.1061/(asce)0733-9445(2000)126:10(1222)

LI Q, MAO J H, HUANG M F. Extreme wind speed estimation of mixed climate combined with typhoon full path simulation[J]. Journal of Vibration and Shock, 2020, 39(23): 84−89(in Chinese). doi: 10.13465/j.cnki.jvs.2020.23.013

FANG G S, PANG W, ZHAO L, et al. Estimation and mapping of extreme wind speeds in southeast coastal region of China[J]. China Civil Engineering Journal, 2021, 54(7): 43−53(in Chinese). doi: 10.15951/j.tmgcxb.2021.07.009

JING R Z, LIN N. Tropical cyclone intensity evolution modeled as a dependent hidden Markov process[J]. Journal of Climate, 2019, 32(22): 7837−7855. doi: 10.1175/jcli-d-19-0027.1

WANG K F, GOU C, DUAN Y J, et al. Generative adversarial networks: The state of the art and beyond[J]. Acta Automatica Sinica, 2017, 43(3): 321−332(in Chinese). doi: 10.16383/j.aas.2017.y000003

KOZAR M E, MISRA V, POWELL M D. Hindcasts of integrated kinetic energy in Atlantic tropical cyclones: A neural network prediction scheme[J]. Monthly Weather Review, 144(12): 4591-4603. doi: 10.1175/mwr-d-16-0030.1

GAO S, ZHAO P, PAN B, et al. A nowcasting model for the prediction of typhoon tracks based on a long short term memory neural network[J]. Acta Oceanologica Sinica, 2018, 37(5): 8−12. doi: 10.1007/s13131-018-1219-z

HUANG C, BAI C, CHAN S X, et al. MMSTN: A multi-modal spatial-temporal network for tropical cyclone short-term prediction[J]. Geophysical Research Letters, 2022, 49(4): e2021GL096898. doi: 10.1029/2021gl096898

LI X Q, YANG W L, LI N N. Cross-language sentiment classification method based on adversarial bidirectional gate recurrent unit network[J]. Computer Applications and Software, 2024, 41(1): 82−88(in Chinese).

NI Y H, LU H, JI C, et al. Comparative analysis on bridge corrosion damage detection based on semantic segmentation[J]. Journal of Southeast University (Natural Science Edition), 2023, 53(2): 201−209(in Chinese).

XIONG J C, CHEN J. A generative adversarial network model for simulating various types of human-induced loads[J]. International Journal of Structural Stability and Dynamics, 2019, 19(8): 1950092. doi: 10.1142/s0219455419500925

WU F B, HUANG G Q, ZHOU X H. Enhanced circular subregion method in typhoon hazard analysis[J]. Journal of Structural Engineering, 2021, 147(6): 06021003. doi: 10.1061/(asce)st.1943-541x.0003003

CHEN Y, YANG J, DUAN Z D, et al. Typhoon hazard analysis of the Guangdong-Hong Kong-Macao greater bay area[J]. Journal of Natural Disasters, 2022, 31(2): 26−38(in Chinese). doi: 10.13577/j.jnd.2022.0203

DARLING R W R. Estimating probabilities of hurricane wind speeds using a large-scale empirical model[J]. Journal of Climate, 1991, 4(10): 1035−1046. doi: 10.1175/1520-0442(1991)0041035:epohws2.0.co;2

VICKERY P J, WADHERA D. Statistical models of Holland pressure profile parameter and radius to maximum winds of hurricanes from flight-level pressure and H*Wind data[J]. Journal of Applied Meteorology and Climatology, 2008, 47(10): 2497−2517. doi: 10.1175/2008jamc1837.1

HONG X, WAN Z Q, CHEN J B. Parallel assessment of the tropical cyclone wind hazard at multiple locations using the probability density evolution method integrated with the change of probability measure[J]. Reliability Engineering System Safety, 2023, 237: 109351. doi: 10.1016/j.ress.2023.109351

LU X Q, YU H, YING M, et al. Western North Pacific tropical cyclone database created by the China meteorological administration[J]. Advances in Atmospheric Sciences, 2021, 38(4): 690−699. doi: 10.1007/s00376-020-0211-7

ISHII M, SHOUJI A, SUGIMOTO S, et al. Objective analyses of sea-surface temperature and marine meteorological variables for the 20th century using ICOADS and the Kobe Collection[J]. International Journal of Climatology, 2005, 25(7): 865−879. doi: 10.1002/joc.1169

COMPO G P, WHITAKER J S, SARDESHMUKH P D, et al. The twentieth century reanalysis project[J]. Quarterly Journal of the Royal Meteorological Society, 2011, 137(654): 1−28. doi: 10.1002/qj.776

MONTEREY G I, LEVITUS S. Climatological cycle of mixed layer depth in the world ocean [R]. U. S. Gov. Printing Office, NOAA NESDIS, 1997.

LI S H, HONG H P. Typhoon wind hazard estimation for China using an empirical track model[J]. Natural Hazards, 2016, 82(2): 1009−1029. doi: 10.1007/s11069-016-2231-2

CHEN Y, DUAN Z D. A statistical dynamics track model of tropical cyclones for assessing typhoon wind hazard in the coast of Southeast China[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2018, 172: 325−340. doi: 10.1016/j.jweia.2017.11.014

中华人民共和国住房和城乡建设部. GB50009-2012, 建筑结构荷载规范[S]. 北京: 中国建筑工业出版社, 2012.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于深度学习的非定常周期性流动预测方法

个体间距对多扑翼气动噪声的影响特性

安装效应对螺旋桨纯音噪声的影响

基于Ac-准则的最优输入设计方法

大展弦比飞行器突风响应数值计算